Go to:     Open Ergonomics Home     PeopleSize Zone      Safe Office Practice

Psz Home Page     Comparison of the 2 Versions     Prices & How to Buy     Details      PeopleSize FAQ

Technical Document: Populations

[<< Previous Chapter] [Chapter Index] [NEXT CHAPTER >>]

   1. Adult Data
   2. Child Data


1. Adult Data
British
Chinese
Dutch
French
German
Italian
Japanese
Loughborough Students
USA

British

The best data exists only for England. Stature and weight data from carefully sampled health surveys in 1994 and 1995 were averaged and weighted according to the national age distribution (refs 77, 78). Together these surveys give a sample size between 13,678 and 16,443. The sampling closely matched the national social class distribution.

In 1993 the UK population was distributed (ref 79):
83 % in England
3 %in Northern Ireland
9 % in Scotland
5 % in Wales

The HUMAG dataset was analysed to determine whether the English data could be used to represent the whole British/UK population. Although the Welsh, Scottish and Northern Irish populations are slightly shorter and lighter, the effect is reduced to a negligible level by the small proportions. It was found that using the English data over-estimates stature by 0.5 to 2mm and weight by up to 0.2 kilogrammes. Since the evidence suggests that both these dimensions are increasing over time, the English data are probably the best possible estimates of the UK population for the end of the 20th Century.

Because the weights were measured in light indoor clothing, a small adjustment had to be made. Measurements indicated that 750 grammes is an appropriate reduction for adults. In order to scale clothing weight for size, all gross weights were reduced by 1%.

Back To Top


Chinese

The Chinese statures were calculated as weighted averages of data supplied by Professor L. Lim of Nanyang Technological University, Singapore (ref 82) and reported by Li (ref 10). The Singapore data is for workers, chiefly in manufacturing, and the Taiwanese study is of civilians from unspecified occupations. The two are felt to provide a good estimate of urban Chinese data (rural Chinese are likely to be smaller).

Back To Top


Dutch

The Dutch statures were taken from data kindly supplied by Molenbroek, of Delft University of Technology (ref 52). 265 male students and 89 female were measured at 4 different times between 1985 to 1986. In 1986, 152 males and 457 females aged 65-74 were measured.

Back To Top


French

Ratios with stature and weight were obtained for the dimensions measured by Rebiffe (ref 86), 1981-1982. The ratios were validated against Haslegrave, C.M. study (ref. 32) and then updated with statures and weights by using averages of relevant age groups from the 1991 INSEE Premiere publication (ref 87). These were self-reported data, but the best we have been able to find. The sample size was over 21,000. The statures validate well from Rebiffe and the Insee 1980 survey in relation to the change in UK statures from Haslegrave 1981 to the Department of Health 1990's surveys. The weights have not increased, and there is no way of telling whether this is optimism on the part of respondents or a real cultural difference. Notwithstanding this uncertainty, if you are designing for French adults these are the best data available.

Back To Top


German

The data are from the German National Health Survey 1990-1992 (ref. 88). This was a representative survey in Germany, including the age groups 25-69 years and persons with German nationality only. Measurement of height and weight was carried out with standarized measures, the study subject had to remove the outer garments and their shoes.

Back To Top


Italian

The Italian statures are as reported in data supplied by Masali, of the University of Torino (ref. 89). Italians were measured on the beaches near Ancona in 1990. The majority of these subjects were from the highly industrialised regions of Northern Italy so a second survey was conducted in 1991 to obtain measures from Southern Italians. The two regions are reported separately as the Italians in the North & Central regions are both taller and heavier than their Southern counterparts.

Back To Top


Japanese

The Japanese statures were calculated as a weighted average of data for the relevant ages as reported by Kimura (ref 80) and from 1992 studies undertaken by the Research Institute of Human Engineering for Quality of Life (HQL). An additional source of female data was supplied by Kumi Ashizawa (ref 81).

Younger Japanese people have leg/trunk proportions much closer to Caucasian populations than used to be the case (Tanner 1982), and are much taller than older Japanese people. PeopleSize version 1 used a narrower age range (18 to 35 years) and the mean statures were 1691mm for men and 1562 for women. The small reduction seen below is due to the wider age range.

Back To Top


Loughborough Students

Measured in 1996 & 7 as part of the ergonomics and design courses at Loughborough University in the UK. Loughborough is a noted sports University. First Year students, including a few mature students, were measured, with a total sample of 300. Distinctly taller than students in the national surveys.

Back To Top


USA

The National Health and Nutrition Examination Survey (NHANESIII), 1988-94 was used to calculated stature and weight data for the different age groups. The raw data were weighted to ensure results are representative of the US population.

Back To Top


2. Child Data

Fewer sources were found for children than adults and some dimensions are ratios from one source, rather than an average of several.

Infant data (birth to 24 months) are from the HUMAG Research Group (at Loughborough University) who undertook a survey in 1978 in which 2,000 British infants were measured. The weights and lengths of these infants have not been updated to the current decade.

British child statures and weights (2 to 17 years) are from the Department of Health 'Health Survey for England 1995'.

US child statures and weights (2 to 17 years) are from the National Center for Health Statistics, NHANES III, 1988-94. US children tended to be slightly taller and heavier than British children of the same age, but these differences were smaller than expected.

The HUMAG Research Group also measured 2000 children aged between 4 and 6 years and 10,000 boys and girls aged from 7 to 16 years. The data have been brought up to date using the technique described in Estimation Methods.

Other sources of data used were Snyder et al 1977 and DIN 33 402. No up to date stature or weight data were found for German children.

The child data were smoothed to show appropriate growth from year to year for each dimension. This is frequently necessary where cross-sectional studies are used and where sample numbers become small in one or more year groups.

Standard deviations were capped where necessary to limit the Coefficients of Variation (SD/Mean*100). Standard deviations tend to be too big in small samples, and child data often suffer from this problem because of the division of the total sample into 15 age groups.

Back To Top

[<< Previous Chapter] [Chapter Index] [NEXT CHAPTER >>]

Psz Home Page     Comparison of the 2 Versions     Prices & How to Buy     Details      PeopleSize FAQ

 

  Open Ergonomics Home Page
 

© Open Ergonomics Ltd.